In radio systems implemented by CDMA (Code Division Multiple Access), the operation is based on scattered spectrum communication. The transmitted signal is scattered by an individual hash code Wi, e.g. a Walsh code, addressed to the subscriber, whereby the transmission is spread out on a broadband radio channel, which is e.g. 1,25; 6,4 or 20 MHz. Hereby it is possible on the same broadband radio channel to transmit simultaneously to several subscribers CDMA signals processed with different hash codes. In CDMA systems, the special hash code of each subscriber hereby produces a traffic channel in the system in the same sense as the time slot in a TDMA system. E.g. in an IS-95, the hash code means a code, by which the channels are distinguished from one another. All transmitted hash codes can also be multiplied by a scrambling code which is specific for the base transceiver station or the antenna. At the receiving end, the CDMA signal is broken up using the subscriber's hash code, whereby a narrow-band signal will result. The broadband signals of other subscribers represent noise in the receiver beside the desired signal. In this application reference is mostly made to a CDMA system in accordance with the IS-95 standard, although the invention is suitable for any system. In CDMA systems, the communication may be TDD-based (Time Division Duplex), whereby the uplink and downlink directions of the base station connections are set up at the same frequency in different time slots, or FDD-based (Frequency Division Duplex), whereby the frequencies of the uplink and downlink channels differ from each other by a duplex frequency.
It must be possible in all cellular systems to regulate at least the transmission power of the mobile station, so that its transmission will arrive at the base transceiver station with an adequate signal-to-noise ratio irrespectively of how far the mobile station is located from the base transceiver station. Power control is described in the following using a CDMA system as an example. FIG. 1 in the appended drawing shows a downlink CDMA traffic channel (Forward Traffic Channel), which includes the following code channels: a common channel, one synchronising channel, from one to seven calling channels and a maximum number of 61 traffic channels. The maximum is when there is only one calling channel besides the synchronizing channel. Each code channel is orthogonally spread and thus broadened by using a random noise sequence phase quadrature pair.
On the common channel a scattered spectrum signal is transmitted constantly at standard power, which signal is used for synchronising the MS mobile stations and for other broadcasting to the mobile stations.
The traffic is used for transfer of user and signalling information to the mobile station MS. Each downlink traffic channel contains a power control sub-channel, on which power control commands are transferred to the mobile station during the connection, in response to which the mobile station will change its transmission power.
The power control sub-channel is formed so that power control bits are contantly transmitted among the bits of the normal traffic channel. The bits are located in the frame so that from the completed traffic channel, which is a convolution coded and interleaved frame formed by modulation symbols, two successive modulation symbols are removed at regular intervals and they are replaced by power control bits. The procedure is generally known in the field and it is called symbol puncturing. The puncturing pattern shows which symbols are removed from the frame and replaced with power control bits.
FIG. 2 of the appended drawing shows signal transmission from the sender to the receiver in a telecommunications system. The information to be transmitted is conveyed over a transmission channel, such as a radio channel, modulated in a form suitable for the channel. Non-idealities of the transmission channel, such as signal reflections, noise and interference caused by other connections, will cause changes in the signal containing the information, so the signal perceived by the receiver is never an exact copy of the signal transmitted by the sender. By using channel coding the information transmitted in digital systems can be made to tolerate better any non-idealities of the transmission path. At the receiving end, the receiver corrects the received signal in a channel corrector based on a channel estimate, that is, with the aid of the channel characteristics known to itself, and it undoes the modulation and channel coding used on the transmission channel.
The mobile station MS is usually connected to a base station providing the best signal quality. For the duration of a handover during the call the mobile station may in CDMA systems be simultaneously connected to several base stations BS, until some base station signal turns out to be better than the others, whereby the call is continued through this base station BS. A handover of this kind is called soft handover.
In a mobile communications system environment, the reliable transmission is disturbed by signal fading on the radio path. To compensate for fading, various means have been developed in mobile communications systems, such as reception diversity and power control as well as some solutions based on diversity use of the transmission antenna. The published patent application EP-741 465 presents one such transmission diversity solution for implementation at the base station. In the publication, the mobile station chooses the best signal out of several transmission antenna signals and notifies the base station of this choice, whereupon the base station continues the transmission through this chosen antenna. The base station adds a first packet identifier to the first data packet and transmits the first data packet together with its identifier through one antenna. Correspondingly, the base station adds the second packet identifier to the second data packet and sends the second data packet to packet together with its identifier through another antenna. The mobile station receives both these transmissions and compares the received signal levels with one another. Having chosen the optimum transmission branch, the mobile station notifies the base station of the chosen packet identifier in a control time slot. The base station sends the transmission intended for the concerned mobile station by way of the indicated antenna. According to the publication, the signals of several signal connections are combined in the coding stage before the transmission is directed to different transmission branches, so when using the antenna diversity according to the publication, all user data is sent through the same chosen transmission antenna branch. The method according to the publication is suitable for use also when the transmission and reception frequencies are different, that is, when frequency-division duplexing FDD is used.
It is a problem with known transmission antenna diversity methods and with the solution of the EP publication described above that the choice of antenna is centralised jointly for all subscriber connections, whereby all traffic is directed for transmission by way of one antenna. Thus, the solution of the EP publication is not suitable for setting up several simultaneous subscriber connections by way of different antennas. It is another problem that the antenna chosen by a mobile station may be misinterpreted at the base station, when the choice notification is based on information of one bit or a few bits in the individual message. Due to data transmission errors, this information may be erroneous in the reception. If the base station misinterprets the antenna chosen by the mobile station and then transmits by way of the misinterpreted antenna, the data transmission quality will suffer from the base station to the mobile station, as the mobile station assumes that the transmission comes from the antenna it has chosen. The waiting value of the false interpretation in the information may hereby be up to 0.5. If the choice antenna is misinterpreted at the base station, then e.g. the interpretation of power control bits may fail in the downlink direction. A further problem with the EP publication described above is that channels set up through different antennas are not orthogonal in relation to one another, whereby they will cause interference to each other.